Recombinant human erythropoietin reduces plasminogen activator inhibitor and ameliorates pro-inflammatory responses following trauma.

BACKGROUND AND THE PURPOSE OF THE STUDY: sBesides its hematopoietic effects, erythropoietin (EPO) by mobilization of iron and modulation of some inflammatory cytokines has antioxidant and anti-inflammatory properties. The purpose of this study was to evaluate these effects of erythropoietin and its impact on organ function in traumatized patients.
METHODS: Twenty-six ICU-admitted traumatized patients within 24 hrs after trauma were randomly assigned to the EPO (received EPO, 300 units/Kg/day) and Control (not received EPO) groups. The inflammatory biomarkers including Tumor Necrosis Factor alpha (TNF-α), Interleukin 1 (IL-1), Plasminogen Activator Inhibitor 1 (PAI-1) and Nitrotyrosine were recorded at the admission, 3, 6 and 9 days thereafter. Acute Physiology and Chronic Health Evaluation (APACHE II) and Sequential Organ Failure Assessment (SOFA) scores were also recorded.
RESULTS: Among 12 patients (EPO group) TNF-α level at the day of 9 (P=0.046), and within EPO group at the days of 3 (P=0.026 ameliorate), 6 (P=0.016), and 9 (P=0.052) were significantly lowered. Level of IL-1 and PAI-1 decreased significantly at days of 3, 6 and 9 post intervention. Also there were significant differences between two groups in the SOFA score during three measured time intervals (the first, third and seventh days).
CONCLUSION: From the results of this study it seems that injection of erythrocyte stimulating agent is well tolerated and inhibits the inflammatory response and oxidative stress following trauma.

RCT Entities:   Population Interventions Outcomes

INTRODUCTION

According to WHO estimation, traumatic injuries are leading cause of morbidity and mortality throughout the world especially in underdeveloped countries (1). Among different reasons and mechanisms, volume loss and hypovolemic shock secondary to hemorrhage is one of the major causes of death of traumatized patients (1). Although fluid therapy and blood transfusion play an essential role in correction of blood loss but, transfusion has its drawback such as infection transmission, acid-base disequilibrium, hypothermia, electrolyte disturbance, deep vein thrombosis, and also decrease patient survival. Blood loss, nutritional deficiency, and endocrine, renal and hepatic insufficiencies are some of the reasons for anemia in traumatized and critically ill patients. Erythropoietin a secreted glycoprotein by kidney through augmenting of hematopoietic remarkably improve anemia and blood oxygen supply in critically ill patients (2).

Production and release of free radicals following traumatic injury is one of the underlying mechanisms leading to cell injury and subsequent organ failure and it has been reported that excess free radicals overwhelmed antioxidant defense of body. Many research have been performed with antioxidant drugs including N-acetylcysteine (3), vitamin C (4), vitamin E (5) and even iron chelators (6) for prevention of oxidative stress following trauma, but results are not convincing. In addition to the obvious role of EPO for treatment of anemia, several clinical reports have shown that it has also antioxidant effects (7).

EPO by increasing hemoxygenase-1 within cells (7), degrade hem and release iron for subsequent utilization in hematopoisis. Metal ions especially iron by production of powerful hydroxyl radicals play an important role in oxidative stress process and it seems that EPO mobilizes iron in blood cell production cycle and decrease oxidative stress. Moreover, increasing numbers of circulating young red blood cells indirectly reduce cellular oxidative stress (7). This study assesses the antioxidant and anti-inflammatory effects of EPO on key systemic inflammatory response syndrome (SIRS) mediators in traumatized patients.

MATERIAL AND METHODS

Study design and setting

This study was an open label randomized clinical trial (Registered under government ID number NCT00622934), conducted at Intensive Care Unit of “Sina” teaching hospital (Tehran University of Medical Sciences, Tehran, Iran). Traumatized patients within 24 hrs after trauma referring to the hospital from October 2008 to October 2009 were enrolled in the study. In all cases, informed consent was obtained from patients or their closest relatives. The study procedure and protocol were approved by the ethical committee of Tehran University of Medical Sciences and Health Services. Randomization was stratified by utilizing a computer-generated random numbers.

Study Population

All patients had been admitted to either a medical, surgical, or a medical/surgical ICU. Inclusion criteria were; staying in the ICU for 3 days, hematocrite less than 28%, and provision of signed informed consent. Exclusion criteria included age less than 18 years, renal failure with dialysis, new onset or uncontrolled seizures, acute burns, pregnancy or lactation, acute ischemic heart disease, acute gastrointestinal bleeding, prior treatment with EPO, participation in another research protocol, and expected ICU discharge within 48 hrs.

Methods

Patients were randomly assigned to two groups. In the EPO group (n=12), 300unit/kg of EPO (Pooyesh Darou-Poe tin®, Tehran, Iran) were injected subcutaneously and continued every other day for a total of 3 doses (days of 1, 3 and 5) and followed up for 10 days. Blood samples were collected from central catheter at the day of admission and 3, 6 and 9 days thereafter. All samples were spun for 15 min and plasma was stored at −80°C until the time of analysis.

The following transfusion guideline was considered for the study: no RBC transfusion if the hemoglobin level was at least 9 g/dl or the hematocrite concentration was at least 27%, unless there was a specific clinical indication (active bleeding, ischemia); RBC transfusion for a hemoglobin level less than 9 g/dl or a hematocrite concentration less than 27% according to physician's judgment.

Patients’ clinical and preclinical characteristics were recorded as follows: consciousness according to the Glasgow Coma Scale (GCS), saturation oxygen tension (SPO2), blood urea nitrogen (BUN), creatinine (Cr), serum electrolytes including sodium, potassium, magnesium, calcium, phosphorus, blood sugar (BS), WBC count, platelet count, hemoglobin, hematocrite, arterial blood gas (ABG), body temperature (using a rectal probe) and blood pressure, pulse rate, respiratory rate 4 times a day or according to the physician's order. All measurements were performed at the admission, within the first 3 days, and at the days of 6 and 9 after admission. Patients also underwent Acute Physiology and Chronic Health Evaluation (APACHE) II scores (8), injury severity score (ISS) (9) based on data obtained within the first 24 hrs after ICU admission. Sequential Organ Failure Assessment (SOFA) (10) as for organ dysfunction was assessed in the first and third and seventh days of admission.

All participants were continuously monitored for their hemodynamic parameters. Following biomarkers were considered as the inflammatory reactants: C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), Tumor Necrosis Factor alpha (TNF-α), interleukin 1 (IL-1), plasminogen activator Plasminogen Activator Inhibitor 1 (PAI-1) and nitro tyrosine test (for anti-oxidant power).

Co-morbidity factors were identified from the medical history including cardiac disease, chronic pulmonary disease, diabetes mellitus, hypertension, malignancy, peripheral vascular diseases, and embolic diseases.

Cytokine measurements

Inflammatory biomarkers concentrations in plasma were measured by using commercially available enzyme-linked immunosorbent assay kits as follows:

TNF-α (Bender Med Systems, Austria, Europe), IL-1 (Bender Med Systems, Austria, Europe), PAI-1 (American diagnostic, Germany, Europe) and Nitrotyrosine (Hycult biotechnology b. v., Netherlands, Europe).

Statistical analyses

The frequency of all demographic and clinical characteristics was calculated. Independent two sample t test was used to compare the mean differences of the clinical and preclinical variables between and within groups compared to baseline. For non-parametric data, Mann-Whitney U test (a non-parametric two independent samples test) was used. The quantitative results are shown as mean (±standard deviation) or number (percent) if necessary. All statistical calculations were conducted using SPSS ver. 14. P and values <0.05 was considered significant.

RESULTS

Twelve EPO and 13 control patients with the mean age of 39.33 (±23.8) and 32.46 (±13.5) years respectively (P=0.38) underwent intervention. All recruited patients were men. Among 26 recruited patients, one patient was transferred to the neurosurgery department. ICU scores including APACHE II, and ISS scores were not significantly different between two groups [The relevant P values (and 95% confidence interval) were 0.43 (−2.31_−0.04), 0.39 (−4.02_1.64) and 0.41 (−0.63_1.47) respectively] (Table 1).

Table 1

Baseline Characteristics.

	Control group n=13	Epo group n=12	P Value
Age, mean (SD), year	39.33 (23.8)	32.46 (13.5)	0.667
APACHE II score, mean (SD)	19.6	19.7	0.43
ISS	13.5	13.7	0.39
SOFA	9.5	7.1	0.41
Baseline inflammatory factors
TNF-α (pg/ml)	12.32	12.21	0.824
IL1 (pg/ml)	6.43	5.34	0.301
PAI1 (ng/ml)	62.07	65.75	0.520
Nitro tyrosin (10 -9 mol)	10.7	6.15	0.299

SOFA score was assessed in the first and third and seventh days, the patients who did not receive EPO had higher SOFA score (Table 2).

Table 2

Comparison of changes in SOFA scores during da

		Epo group	Control group	P value
SOFA	Day 1	7.1±0.5	9.5±0.7	0.006
	Day3	5.8±0.5	8.8±0.8	0.002
	Day7	5.3±0.5	8.9±0.9	0.001

Patients’ consciousness, body temperature, blood pressure, pulse rate (except for higher pulse rate within EPO group at the 6th day, (P=0.042)), respiratory rate, SPO2, BUN, Cr, serum sodium, potassium, calcium & phosphorus, blood sugar, Hb and Hct were not significantly different between and within groups. The Mann-Whitney test showed similar results, except for Hct which was higher as a trend within the 9th day of ICU staying (P value for the day of 3, 6 and 9 were 0.06, 0.08 and 0.06 respectively).

Mann-Whitney U test showed that TNF-α level between groups at the day of 9 (P=0.046), and its level within EPO group at the days of 3 (P=0.026), 6 (P=0.016), and 9 (P=0.052) were significantly lower (Fig. 1).

Figure 1

Changes in plasma concentrations of TNF-α. There were significant differences in TNF-α level between two groups at the day of 9 (P=0.046), and its level within EPO group at the days of 3 (P=0.026), 6 (P=0.016), and 9 (P=0.052) compared with baseline.

IL-1 was significantly lower at the days of 3 and 6 after treatment in EPO group [P=0.051 (95 CI: −0.006_3.03) and P=0.005 (95 CI: 0.89–4.23)] respectively. Moreover, Mann-Whitney U test indicated that IL-1 level at the day of 9 was also significantly lower in EPO group (P=0.030) (Fig. 2).

Figure 2

Changes in plasma concentrations of IL-1. There were significant differences in IL-1 between two groups at the days of 3 and 6 (P=0.051 and 0.005 respectively) and its level within EPO taken group at the day of 9 compared with baseline (P=0.030).

The relevant independent samples t-test analysis about PAI-1 suggest that PAI-1 levels between groups and within EPO group significantly decreased from the day of 3 (P=0.011) toward the end of treatment (P<0.001). Mann-Whitney U test confirmed the same results (P=0.008 for the day of 3, and P<0.001 for the rest of days) (Fig. 4).

Figure 4

Changes in plasma concentrations of Plasminogen Activator Inhibitor 1 (PAI-1). There was significant decrease in plasma level of (PAI-1) in EPO group at the day of 3, 6 and 9 (P=0.11, <0.001 and <0.001 respectively).

The statistical analysis (independent samples t-test) proposed that nitro tyrosine levels were significantly lower in EPO than EPO Control groups at the days of 6 (P=0.006, 95 CI: 2.10–10.52) and 9 (P=0.007, 95 CI: 2.12–11.21). Applying the Mann-Whitney U test indicated that nitro tyrosine level was lowered significantly not only at the days of 6 and 9 (P value were 0.046 and 0.030 respectively) between groups, but also within EPO group at the same days (P value was 0.035 for the day of 6) (Fig. 3).

Figure 3

Changes in plasma concentrations of Nitro tyrosine. Nitro tyrosine levels were significantly lower in EPO group than Control group at the days of 6 and 9 (P=0.006 and 0.007 respectively)

DISCUSSION

EPO exerts humeral effects by means of its specific receptor (EPO-R) which does not exist in many tissues under normal conditions; but several metabolic stressors (like ischemia, hypoglycemia, etc) rapidly provoke up-regulation of the EPO-R which in turn leads to inhibition of production of pro-inflammatory cytokines like TNF-α, IL-1β, IL-6, IL-8, PAF, etc (11). The cytoprotective mechanism of EPO is not completely understood but a probable explanation has been suggested (11).

Our recent investigations have revealed remarkable results on the role of cytokines such as IL, growth factors and oxidative stress mediators in the pathology or prognosis of many chronic and acute diseases and these issues led us to study the effect of EPO on these variables (12).

In the present study, the ICU scores including APACHE II, ISS showed insignificant discrepancy between the study groups; therefore further comparison of clinical and para-clinical (inflammatory biomarkers) wee made between groups irrespective of the severity of trauma.

Results showed that IL-1 as an inflammatory reactant significantly decreased after administration of EPO for 3 days and this effect lasted to the 9th day after the treatment was started. Concerning the TNF-α, there was a remarkable decrease in the relevant levels after 9 days treatment with EPO in comparison with EPO control group; and TNF-α level also significantly decreased in the days of 3, 6 and 9 after treatment compared to the baseline level in the EPO group.

In agreement with reports that EPO has a variety of direct anti-inflammatory effects (13), results of this study showed that EPO significantly diminished synthesis of anti-inflammatory cytokines. This indicates that the observed inflammatory effects of EPO were not mediated by an increase in the principal regulatory cytokines, but were either cytokine independent or mediated by a different cytokine(s). Inhibition of apoptosis has been suggested as a fundamental mechanism of EPO-mediated tissue protection. However, in an invitro model it was found that EPO rescues only a small number of cytokine-producing cells from apoptosis (data not shown).

The cytokine inhibitory properties of EPO likely represent an important mechanism by which its tissue-protective effects are mediated and evidence suggests that cytokine-mediated cell injury plays an essential role for end-organ damage (14).

In accordance with findings of this study, it seems that PAI-1 was affected strongly and remarkably by EPO administration because its level fell significantly from the third day after treatment and remained low during the study period. PAI-1 is the major physiological inhibitor of tissue type plasminogen activator. PAI-1 is an inhibitor of fibrinolysis and increase in its levels are considered to contribute to the pro-coagulant state associated with thrombotic diseases.

In the past study it was found that PAI-1 level decreased in both intensive insulin therapy and /or combined with metformin (15). Consistent with these results it has been reported that glucose control and insulin treatment improved fibrinolysis by reducing PAI-1 concentration/ activity (16) and impaired fibrinolysis during sepsis has been associated with widespread activation of coagulation through the release of inflammatory cytokines (17).

Nitrotyrosine was significantly lower in the EPO treated patients at the days of 3, 6 and 9 after treatment compared with control patients; in addition its level decreased significantly at the days of 3 and 6 in the EPO group in comparison to their baseline values.

EPO may exert its anti oxidative effects directly by exploiting intracellular anti oxidative mechanisms. Increase in the red blood cells by EPO may also indirectly reduce cellular oxidative stress, as red blood cells are loaded with a substantial amount of antioxidative enzymes.A comparison of SOFA scores on the first, third and seventh day showed that EPO group had lower SOFA score which at least indicate that EPO has anti-inflammatory activity.SOFA score was selected as a quantitative tool to judge this matter to some extent because it represents the dynamics of illness including the effectiveness of therapy.

Among critically ill patients, erythropoietin uses appears to be associated with increased thrombotic events. This is consistent with the results of recent trials involving non- critically ill patients with either cancer or chronic renal failure. In these studies, when erythropoietin, was used to achieve higher target hemoglobin concentrations (i.e.,>120 g/l), the risk of thrombotic complications and death were increased (18, 19). Importantly, patients with past thromboembolic complications had to be strictly excluded, and therefore patients who are prone to bed should not be included in a chronic treatment study. However findings of this study did not show any thromboembolic complications because Platelet counts and other related factors had been carefully and continuously monitored at all times of EPO treatment and even in weeks thereafter. Other side-effects such as hypertension and allergic reactions have been reported (20).

CONCLUSION

Administration of recombinant human erythro-poietin for ICU-admitted traumatized patients seems beneficial as it may helps to restrain the inflammation and oxidative injuries, which in turn may lead to improved prognosis during and after ICU admission.